Skill: Research Summarization

Overview

This skill enables Perplexity AI to rapidly synthesize large volumes of information from multiple sources into concise, structured, and actionable summaries tailored to specific audiences and purposes.

Skill Metadata

• Skill ID: research-summarization
• Category: Knowledge Management & Research
• Complexity: Intermediate
• Estimated Time: 5-30 minutes depending on scope
• Output Format: Structured Summary Document (Markdown)

Capabilities

1. Multi-Source Synthesis

• Aggregate information from web articles, papers, reports, and databases
• Cross-reference facts across multiple sources
• Identify consensus views vs. contested claims
• Highlight primary vs. secondary sources
• Flag information gaps and limitations

2. Audience-Adapted Summarization

• Executive summaries for leadership (300-500 words)
• Technical summaries for practitioners (detailed, jargon-inclusive)
• Layperson summaries for general audiences (simple language)
• Bullet-point briefings for quick consumption
• Narrative summaries for storytelling contexts

3. Structured Output Generation

• Key findings and takeaways
• Supporting evidence and data points
• Conflicting perspectives and nuances
• Implications and recommendations
• Source bibliography and credibility assessment

4. Domain-Specific Research

• Academic and scientific literature
• Market research and industry reports
• Legal and regulatory documents
• News and current events
• Technical documentation

5. Quality Assurance

• Source credibility scoring
• Recency assessment
• Bias identification
• Confidence level tagging
• Fact-check flagging

Input Requirements

yaml
required_inputs:
  - research_topic: "The subject to research and summarize"
  - audience: "Who will read this summary"
  - purpose: "Why this summary is needed"

optional_inputs:
  - scope: "Breadth and depth of coverage"
  - time_range: "How recent the sources should be"
  - length: "brief|standard|comprehensive"
  - format: "bullets|narrative|structured|executive"
  - focus_areas: "Specific aspects to emphasize"
  - output_language: "Language for the summary"

Step-by-Step Workflow

Phase 1: Scope Definition

Step 1: Research Brief Creation

• Define primary research question and sub-questions
• Set geographic or temporal scope
• Identify source type preferences
• Determine depth required and audience expertise level

Step 2: Source Planning

• Primary sources: Original research, official reports, data
• Secondary sources: Analysis, commentary, journalism
• Expert opinions: Interviews, podcasts, talks
• Quantitative data: Statistics, surveys, studies

Phase 2: Information Gathering

Step 3: Systematic Search

Search query templates:

• "Topic] overview 2024"
• "Topic] research findings"
• "Topic] expert opinion"
• "Topic] statistics data"
• "Topic] challenges implications"

Step 4: Source Evaluation

| Criteria | High Quality | Medium Quality | Low Quality | |----------|-------------|----------------|-------------| | Authority | Peer-reviewed, official | Industry reports | Anonymous blogs | | Recency | Last 6 months | Last 2 years | Older than 3 years | | Accuracy | Verified, cited | Generally accurate | Unverified claims | | Bias | Neutral/disclosed | Minor bias | Strong agenda |

Phase 3: Synthesis & Writing

Step 5: Theme Identification

• Cluster related information into themes
• Identify recurring points across sources
• Note divergent perspectives
• Prioritize by relevance to purpose

Step 6: Hierarchical Organization

1. Executive Summary (top-line findings)
2. Key Themes organized by importance
3. Supporting Data and Evidence
4. Conflicting Views if applicable
5. Implications and Recommendations
6. Sources and Credibility Notes

Step 7: Writing for Audience

• C-Suite: Lead with business impact, skip technical details
• Technical teams: Include methodology, data, and specifics
• General public: Define jargon, use analogies, tell stories
• Policy makers: Emphasize implications and action options
• Investors: Focus on opportunity, risk, and financial metrics

Output Templates

Executive Summary Template

# [Topic] Executive Summary
Prepared: [Date] | For: [Audience] | By: [Author]

## Bottom Line Up Front
[2-3 sentences on what you need to know and why it matters]

## Key Findings
1. [Finding 1]: [One sentence with supporting data]
2. [Finding 2]: [One sentence with supporting data]
3. [Finding 3]: [One sentence with supporting data]

## Implications
- Short-term: [Immediate actions or impacts]
- Long-term: [Strategic considerations]

## Recommended Actions
1. [Action 1]
2. [Action 2]

## Sources
[List of key sources used]

Comprehensive Research Summary Template

# Research Summary: [Topic]
Scope: [Geographic/Temporal scope]
Date Range: [Coverage period]
Sources Reviewed: [Number and types]

## Executive Overview
[150-200 word summary]

## Background and Context
[Why this topic matters, historical context]

## Key Themes and Findings

### Theme 1: [Title]
Summary: [2-3 sentence overview]
Evidence:
- [Data point or finding]
- [Supporting example]
- [Expert quote or opinion]
Confidence Level: High/Medium/Low

## Data and Statistics
| Metric | Value | Source | Date |
|--------|-------|--------|------|

## Contrasting Perspectives
Mainstream View: [Summary]
Alternative View: [Summary]
Why it matters: [Implication of the disagreement]

## Gaps and Limitations
- [Information not found or unavailable]
- [Areas needing further research]

## Implications and Recommendations
- [Recommendation 1]
- [Recommendation 2]

## Source Bibliography
| Source | Type | Credibility | Date |
|--------|------|------------|------|

Quality Checklist

• All key claims are supported by at least one source
• Sources are recent within required time range
• Multiple perspectives are represented
• Bias has been identified and noted
• Technical terms are defined for target audience
• Summary length matches requested format
• Confidence levels are assigned to uncertain claims
• Recommendations follow logically from findings

Domain-Specific Adaptations

Scientific/Academic Research

• Prioritize peer-reviewed sources
• Include study methodology and sample sizes
• Note statistical significance
• Distinguish correlation from causation

Market and Industry Research

• Include market size and growth data
• Identify key players and market share
• Note analyst ratings and forecasts
• Highlight investment signals

Legal and Regulatory Research

• Cite specific laws, regulations, and rulings
• Note jurisdiction and applicability
• Highlight compliance requirements
• Flag areas of legal uncertainty

News and Current Events

• Prioritize recency (last 24-72 hours)
• Cross-reference across multiple outlets
• Track story evolution over time

Perplexity-Specific Tips

• Use Focus: Academic for scholarly sources and research papers
• Use Focus: News for current events and breaking developments
• Enable Pro Search for comprehensive multi-source synthesis
• Use Follow-up questions to explore specific themes deeper
• Leverage Spaces to organize ongoing research projects
• Use date filters to control recency of sources
• Request citations explicitly for verifiable output

Automation Use Cases

yaml
recurring_summaries:
  daily:
    - Industry news digest
    - Competitor monitoring brief
  weekly:
    - Market intelligence summary
    - Regulatory update briefing
    - Research paper highlights
  monthly:
    - Industry landscape report
    - Technology trend analysis
  on_demand:
    - Due diligence research
    - Topic deep-dives
    - Crisis information gathering

Success Metrics

• Accuracy rate (fact-checked against sources)
• Reader satisfaction score (audience feedback)
• Time saved vs. manual research (target: 75% reduction)
• Source diversity index (variety of source types)
• Summary utility rate (percentage acted upon)
• Completeness score (key topics covered)